South Africa

Op-ed: Did sexual offences, hijackings and murder really increase respectfully by 117%, 93% and 9% during 2016/17?

By Chris De Kock 11 October 2017

If just for a moment you place yourself in the position of a potential traveller to, investor in and/or rating agent of South Africa and you had just receive these figures in the headline above, released by the Statistician-General of South Africa himself, what would you do? By CHRIS DE KOCK.

Given the figures on sexual offences, hijackings and murder released by the Statistician-General, as a seasoned international traveller I would immediately reconsider and postpone my trip to South Africa, even if it is among the top 10 most beautiful destinations in the world (according to my own view after seeing most of them, SA is still number 1).

If I received these figures together with the facts of the 35 Dutch tourists who were hijacked and robbed within an hour after they arrived at OR Tambo International for a 22-day tour and the Peruvian tourists who were robbed in their rooms at a Magoebaskloof hotel in the very same week, I would even have good reason to claim back my deposit or money paid in advance, because clearly people are not even safe travelling in groups. I am not an investor of note or a rating agent at all, but I can just imagine how it would influence my decisions.

Since this news broke on 28 September I experienced the same views of many fellow South Africans who just don’t know what to believe any more and who are so disillusioned with everything that has occurred over the last five years that they prefer to tune in to their favourite soap opera or music.

There is a “So what? What can we do?” attitude that is shocking. It is strange that Statistician-General Pali Lehohla can be surprised that although crime in general, according to Statistics SA and the SAPS, decreased, feelings of insecurity increase, if he himself releases the kind of increases in the title of this article. Especially if all of Statistics SA’s victim surveys done during his tenure as the Statistician-General indicate that South Africans indeed have the capacity to distinguish between different kinds of crime and have different levels of fear of each.

Shoplifting and theft of personal property will not increase fear levels, but house robbery, carjacking, street robbery and murder definitely will. The shocking events in Nyanga and Marikana in Philippi East, days before and after the Statistician-General’s release, may even generate such high levels of fear that it can trigger a revenge, counter-revenge, vigilante frenzy (and Minister Fikile Mbalula may run out of task teams).

The Statistician-General, National Commissioner (or mostly acting national commissioners), and Minister of Police should now stop underestimating the capacity and logic of South Africans to understand the crime situation. As was provided under an ANC government from 2002 to 2012, they should provide the leadership and support to lead South Africans out of this crime and fear mess. They should also stop confusing the public in their effort to say things are not so bad. The release of the latest (2016/2017) Victims of Crime Survey is an excellent example of this.

The 2016/2017 Statistics SA Victims of Crime Survey was released in the same week that the SAPS crime statistics are usually released. Unfortunately or fortunately, the SAPS crime statistics were for the first time in 15 years not released before the end of September, usually in the last week of September. But also this time the bulk of the Victims of Crime Survey report reads like a crime statistics report and not like a victim survey report. The bulk of it consist of figures of crime and comparisons over the past four financial years as if it is the recorded crime statistics. A victims of crime survey’s primary function is to establish the reporting and under-reporting (the so-called dark figure) of crime and the reasons for under-reporting. The police registration figures of crime are the crimes reported to them and which they can analyse and use as a basis for policing (it is literally the crime that lands on their table and which they and their partners should deal with – as an old police colleague put it to this author years ago: “The police can only deal with crime that they know of!” The author reacted with “… but we should also know the extent of the crime that we don’t know of and if the police can do anything to know of more crime they should do so.”)

This analyst, who analysed surveys at the HSRC (Human Sciences Research Council) for nearly 20 years before he joined the SAPS in 1995 and who then headed and analysed police recorded crime statistics for the rest of his official career till 2013, can think of so many reasons why national crime figures can’t be calculated from the victims of crime surveys to the population that it would take quite a few articles like this to provide the reasons and explain.

Victims of crime surveys and registered crime statistics can never replace each other. We will have to wait until the “mighty” SAPS release the crime statistics before we know if there was really respectively a 9%, 93% and 117% increase in murder, hijacking and sexual offences. “Mighty” because these crime figures are the property of the people of South Africa which they paid for in blood. In the meantime the tourist, investor and ratings agent can use the Statistician-General’s figures, which cost the taxpayer millions to survey, as the reality of SA.

In column 1 of the table beneath, the “crime figures” for 2016/2017, as calculated by Statistics SA on the basis of their 2016/2017 Victims of Crime Survey are provided (see Statistics SA. 2017. Victims of Crime Survey. 2016/2017). These are the figures that they and now other people use in their calculations of the examples in the title of this article.

In column 2 are the number of cases registered by SAPS for 2015/2016. It would have been better to use the 2016/2017 crime statistics but those are long overdue.

Column 3 provide figures which were calculated on the basis of the 2015/2016 crime figures and the of percentage under-reporting as established by Statistics SA in either the 2016/2017 victim survey marked with * or the 2013/2014 victim survey marked with **.



CASES AS CALCULATED BY STATISTICS SA FOR 2016/2017 (On pages 9 & 10 of the 2016/2017 Statistics SA, VOCS Report)


CASES REGISTERED BY SAPS IN 2015/2016 (2016/2017 statistics not released by 10 October)



Murder. 15 256 18 673 21 052**
Sexual Offences. 58 013 51 895 71 678**
Home robbery. 139 073 20 820 36 526*
Hijacking of motor vehicle. 53 003 15 786 15 786**
Robbery (other than home robbery and hijacking). 279 968 130 190 448 931*
Housebreaking or burglary 734 347 325 614 638 459*
Theft of motor vehicle 42 703 53 809 57 244*
Theft out of motor vehicle 129 880 139 386 242 832**
Theft of livestock 157 576 24 715 69 424**

From this table it is clear that there are large differences of up to 73.7% (for home robbery) between the crime figures calculated by Statistics SA for 2016/2017 based on the 2016/2017 victim survey (column 1) and the adapted crime figures in column 3. There are even larger differences up to 85.0% (for home robbery) between the Statistics SA figures in column 1 and the SAPS crime figures in column 2.

There can be no doubt that these different figures for the same crime trends can produce diametrically opposing crime trends – where one analyst finds that a certain crime trend is increasing and another will find that the same trend is decreasing, depending on the data they use. Knowing the SAPS first trimester of 2016/2017 crime statistics, and out of experience the fact that the last quarter could not dramatically change from the first three quarters, I would predict a marginal increase of 1 to 2 percent in murder, a slight decrease in sexual offences of 2-3% and a plus-minus 15% increase in hijacking, which look very different from the increases of respectively 9%, 117% and 93% in murder, sexual offences and hijacking of the Statistician–General.

SAPS must release their crime statistics for 2016/2017 as a matter of urgency. The two departments then owe it to South Africa and its people, who carry the scars of crime and pay taxes for the crime surveys and the registration of crime, to meet and sort out the confusion and differences between the different sets of data. They should then explain to everybody how their data should be used. DM

Dr Chris De Kock is an independent analyst: crime, violence and crowd behaviour.

Photo: Residents of Ocean View burn tyres and blockade the Kommetjie road in protest against the lack of policing in their area in Cape Town, 21 September 2017. Ocean View has been plagued by gang violence. Recently a nine-year-old boy, Aqeel Davids, was killed in suspected gang crossfire. Residents claim the police and the ward councillors responsible for the area not doing enough to combat the problem while innocent residents are being killed. Photo: Nic Bothma/(EPA-EFE)



Victim, villain or pawn: Why Eskom ‘whistle-blower’ Suzanne Daniels was fired

By Jessica Bezuidenhout